Overview of the notion that cognitive processes are deeply rooted in the body’s interaction with the world. Specific claims examined:

  • Cognition is situated - it takes place in a real world environment and is inherently contextual.

  • Cognition is time pressured - the limits imposed by the real-time nature of cognition inherently shape the process.

  • We offload cognitive work onto the environment - due to resource limitations, we offload cognition to the environment.

  • The environment is part of the cognitive system - the link between mind and environment is so high-bandwidth that it doesn’t make sense to study the mind separate from the environment (note: this claim is described as “deeply problematic” in the abstract).

  • Cognition is for action - if cognition is for controlling our biological body, it must be understood in the context of performing actions.

  • Offline cognition is body based - because the mind evolved contextually with the environment, even “abstract” thought hijacks sensorimotor facilities for cognition.

Notes there are (at least) two schools in the philosophy of mind/cognitive sciences: the mind is an abstract information processor and sensorimotor inputs/outputs are peripheral. Alternatively, brains are control systems for biological bodies and thus sensorimotor processing is intimately tied to cognition.

A deeper examination of the claims:

Cognition is situated

A distinction is drawn between situated and unsituated cognition. The former is the cognition that happens as you do a task (e.g. hunting an animal), the latter is the cognition that happens when you plan or daydream. These might be variations on a theme (i.e. see the body-based cognition claim).

The author concludes that the argument that all important cognition is situated (or rooted in situatedness) isn’t strongly persuasive. I feel a lot of this is splitting hairs on definitions of particular technical terms.

Cognition is time pressured

This claim is, roughly, that the time pressure involved in situated cognition is one of the primary shaping forces of cognition. Brains evolved to, for example, help us flee predators as they jump out of bushes and this deeply influences how evolution shaped our cognitive processes.

Definition: representational bottleneck - time pressure causes a bottleneck which forces us to rely on heuristic tricks

To temper this claim, a lot of what we identify as human cognition is not aggressively time constrained. For example, we do some of our best, recognizably-human work when we can sit and reflect and plan.

It is noted that spatiotemporal and movement behaviors are probably deeply influenced by time constraints. However, the author is unpersuaded that this can be “scaled up” and applied usefully to all of cognition.

We offload cognitive work to the environment

This is using the environment to work around real-time representational bottlenecks. This is distinct from offloading to prevent memorizing (e.g. using a calendar) although they are sometimes conflated in the literature.

This, for example, could be moving furniture around a room to see how it looks rather than trying to imagine how it would look. This is also referred to as “using the world as its own best model”.

The author notes that it seems we use the environment for both time-pressured situated tasks as well as unsituated tasks and that this may have far reaching consequences for how we think about cognition.

The environment is part of the cognitive system

A stronger claim than previous; if we use the environment to offload processing then the environment should always be considered a part of cognition.

Alternatively, directly from the paper: “The claim is this: The forces that drive cognitive activity do not reside solely inside the head of the individual, but instead are distributed across the individual and the situation as they interact. Therefore, to understand cognition we must study the situation and the situated cognizer together as a single, unified system.”

The author claims that causes of cognition are undeniably environmental and internal, but that shouldn’t necessarily constrain the examination of cognition as a modular process. The issue ends up being “how best to carve nature at its joints” and depends deeply on what you’re studying and what you’re trying to get out of the study.

The author gives the example of hydrogen from chemistry. Before there was a theory of an atom, much was known about how hydrogen reacts in many different situations. As atomic theory was developed though, so was a useful understanding of hydrogen as something unto itself. Both lines of research were fertile.

Cognition is for action

The proposal here is that memory shouldn’t be viewed as a tool for recording abstract information, but rather a tool for encoding patterns of possible action in a 4D world.

This claim is tempered by the fact that we remember things that we cannot directly interact with (e.g. sunsets are mentioned, even though we do look at them). More broadly, there is suggestion that there different types of memory. Some are more closely coupled with action and others are more abstract.

The other tempering fact is that it seems we often build up richer representation of objects beyond the here-and-now usefulness. For example, the piano that is a musical instrument that becomes firewood once the blizzard strikes. The representations allow us to be creative when the environment changes.

Offline cognition is body based

The claim is that we hijack our sensorimotor tools to do other, higher-levels of cognition. Examples include mental imagery, different types of sensory-linked working memory, episodic memories, etc.

Excerpts from the paper’s conclusion:

  • “One benefit of greater specificity [of the claims around embodied cognition] is the ability to distinguish on-line aspects of embodied cognition from off-line aspects.”

  • In online, situated cases of cognition, “the mind can be seen as operating to serve the needs of a body interacting with a real-world situation.” Issues of time pressure may come into play. We should be careful, however, in scaling up the processes that come into play to explain all of cognition.

  • In offline, unsituated forms of cognition, “rather than the mind operating to serve the body, we find the body (or its control systems) serving the mind. This takeover by the mind, and the concomitant ability to mentally represent what is distant in time or space, may have been one of the driving forces behind the runaway train of human intelligence that separated us from other hominids.”

THOUGHTS: a decent survey of the field, gives good high-level pointers. Sometimes a little overly-hair-splitting when it comes to definitions.